A common complication after abdominal surgery and, in some cases chest surgery, is that the patients' intestines immediately post-surgery do not function properly. Such patients are often afflicted with "paralytic ileus" and are at increased risk for emesis, aspiration and secondary pneumonias. They are more prone to surgical complications because they fail to properly empty irritating gastric acid from their stomachs and intestinal fluid from their bowels. For example, wound disruption may occur because of the strain of vomiting the gastric acid. These patients may not be permitted to ingest food for several days after their operations. In such cases, nasogastric intubation is often necessary to achieve adequate gastric emptying. Nourishment cannot be provided by the gastrointestinal route during this period. Wound healing and the immune response are therefore depressed in these patients. Nutrition, if provided, requires intravenous feeding which has its own set of risks.
Nasogastric intubation can be extremely uncomfortable for the patient. Further, the procedure itself can cause medical complications. However, delaying the reinstitution of nutritional support in patients is an undesirable alternative because such patients have an increased metabolic demand while their bodies attempt to heal from surgery. A need clearly exists for a method to efficiently and safely stimulate the gastrointestinal tract such that patients may receive nourishment and have a more normal gastrointestinal function in the immediate post-surgical recovery period.
Medical personnel have used antacids and H2 blockers to neutralize the gastric acid which accumulates during delayed gastric emptying, but these treatments can incur a significant expense and do not efficiently remedy the underlying problem. Bowel stimulatory agents, such as metaclopramide, are often used but are not very effective.
Another possible avenue of treatment is direct treatment of the inactive organ system. Over the last century researchers have directed considerable effort towards the study of gastrointestinal motility. The duodenal bulb has been postulated to be the anatomic site of the duodenal pacemaker and as such would be the area responsible for generating activity fronts. Y. Ruckebusch and L. Bueno, Gastroenterology, 72:1309-14 (1977). The aborad propagation of activity fronts is known to .be dependent on enteric cholinergic neurons. S. K. Sarna, et al., Am. J. Physiol., 241:G16-23 (1981). The central nervous system does not initiate migrating motor complex ("MMC") activity in the intestines but does modulate this activity. In the human MMC, the site of initiation is not constant and may begin at sites other than the proximal bowel. MMC rapidly moves interdigestive contents down the small intestine thereby clearing the small intestine of residual food, secretions, and desquamated cells, preventing stagnation and bacterial growth. M. H. Sleisenger and J. S. Fordtran (eds.), Gastrointestinal Disease: Pathophysiology, Diagnosis, Management, Vol. II, "Movement of the Small and Large Intestine," 1088-1105 (1989).
Much of the interest in mechanisms of gastrointestinal motility stems from a desire to regulate intestinal motor activity in pathologic situations such as paralytic or adynamic ileus. Early treatments included the use of electric current as a regulator. F. Katona et al., Wiener Klinische Wochenschrift, 71:818 (1959) (hereinafter "Katona et al."). Subsequent studies of the adynamic gastrointestinal tract attempted to apply electrical stimuli that were strong enough to contract "paralytic" intestinal segments that retained reflex irritability. A. M. Bilgutay and C. W. Lillehei, Ann. Surg., 158:338 (1963); A. M. Bilgutay et al., University of Minnesota Medical Bull., 36:70 (1964) (hereinafter "Bilgutay et al.").
Bilgutay et al. reported that electrical stimulation of the antrum in post-operative patients reduced the average time to first reported flatus from 55 hours to 20 hours. This desirable clinical effect, however, could not be confirmed in subsequent studies. D. C. Quast et al., Surg. Gynec. Obstet., 120:35-37 (1965); J. M. Moran and D. C. Nabseth, Arch. Surg., 91:449-451 (1965); T. Berger et al., Nord. Med., 74:1031 (1965); C. Wells et al., Lancet, I:4-10 (Jan. 4, 1964). In 1966 Sonneland concurred with other prominent investigators in the field that the electronic gastrointestinal pacemaker was of no value. J. Sonneland, Am. J. Surg., 111:200-201 (1966).
In 1977 Gladen and Kelly reported that the canine gastric and intestinal pacesetter potentials were different as shown by electrical pacing. This finding helped to explain why electric stimulation of the gastric pacesetter would not regulate the intestinal pacesetter potential or alter intestinal movement. H. E. Gladden and K. A. Kelly, Mayo Clin. Proc., 52:51-53 (1977). More recent attempts to electrically pace the human intestine post-operatively have likewise met with failure. Soper et al., Surgery, 107:63-68 (1990), reported that electric pulses (50 msec, 5 to 15 mA, 11-13 cpm) did not alter or regulate the pattern of intestinal pacer potentials in any patient at any time after operation.
Clearly, there has been a long and unmet need for effective treatment of gastrointestinal disorders following surgery. In spite of numerous attempts to address this problem with direct electrical stimulation, the loss of normal gastrointestinal function following many surgical procedures remains a serious clinical problem.
The present inventors have discovered that chemical stimulation is an alternative to electrical stimulation for enhancing contractility in the adynamic gastrointestinal tract. Although local chemical stimulation of the gastrointestinal tract had not heretofore been used in an in vivo setting, certain chemicals have been shown to influence the motor activity of bowel tissues. The effect of muscarinic agonists such as acetylcholine on isolated intestinal smooth muscle preparations has been characterized. For example, concentrations of acetylcholine in the range of 10.sup.-6 to 10.sup.-8 g/ml have been shown to cause contraction of such ex vivo preparations. C. F. Code (Ed.) Handbook of Physiology, Vol. IV, "The Alimentary Canal, Motility," 2173-2187 (1968) (hereinafter "Handbook of Physiology"). In muscle preparations that spontaneously contract and relax, this activity is either made more frequent or is abolished and replaced by a steady high level contraction. Of note is the fact that depolarized smooth muscle contracts when exposed to acetylcholine. This suggests that neither the normal resting membrane potential nor the discharge of action potentials is required for the action of the agent, if the action sought is muscle contraction. Handbook of Physiology, 2173-2187.
The study of chemical action on intestinal smooth muscle is complicated by the presence of viable neurons even in isolated preparations. Consequently it can be difficult to determine whether the effect of a given chemical is the result of a direct effect on the smooth muscle or whether the effect is mediated by neuronal elements. Katona et al. Under normal conditions acetylcholine is released from certain nerve endings. If the concentration of the transmitter reaches a threshold level, depolarization of the muscle membrane occurs. Handbook of Physiology, 2173-2187.
Controlled release systems have distinct advantages over conventional drug therapies. Conventional systemic therapies go through a cycle in which the concentration of the drug steadily increases upon ingestion of the drug, peaks and subsequently declines over time. Each drug has an upper threshold over which the drug is toxic to the patient, and a lower threshold under which it is not effective. Repeated cycling can induce periods of alternating toxicity and ineffectiveness. Further, in conventional therapies, the therapeutic agent circulates through the patient's bloodstream, coming into contact with numerous organ systems. Often a high systemic dose is required so that the targeted organ system receives an adequate dose of the drug. An advantage, therefore, of a controlled, localized delivery system is that the drug is targeted to a selected organ system, thereby minimizing adverse side effects caused by either high systemic dosages or presence of the drug in other organs. Controlled drug release systems have been used to treat brain disorders, such as Parkinson's disease. U.S. Pat. No. 4,883,666.
For the purposes of drug delivery, selected drugs have been combined with biocompatible polymer matrices. For example, silicone elastomers may be used as permeant carriers. Their use in the health care field has increased in recent years because of the relative high permeability of silicone elastomers compared with other elastomers. The high rate of permeation is due to the flexible nature of siloxane polymer chains. This increased mobility effectively increases the usable volume of the polymer.
Several types of drug delivery systems using silicone elastomers have been described in the literature. The most commonly used method depends on release of drug through the walls of a hollow device, as in a capsule or tubing. P. J. Dzuik and B. Cook, Endocrin., 28:208 (1966); S. J. Segal and H. Cook, 23rd Meeting of Amer. Fert. Soc., Apr. 14-16, 1967, Washington, D.C.; J. Folkman and V. H. Mark, Trans, N.Y, Acad, Sci., 30:1187 (1968); E. R. Garrett and P. B. Chemburkar, J. Pharm. Sci., 57:1401 (1968); F. A. Kincl et al., Acta Endo., 64:253 (1970); K. Sundaram and F. A. Kincl, Steroids, 12:517 (1968); H. Tatum, Contracep., 1:253 (1970); W. E. Berndton et al., Endocrin., 62:1 (1974).
A second type of drug delivery device consists of a mixture of silicone elastomer and the powdered form of a drug. The device is catalyzed and then cast into a desired shape. A number of drugs have been investigated in this regard and their release from the drug-polymer matrix documented. G. L. Neil et al., Chemother., 18:27 (1973); J. Folkman and D. M. Long Ann. N.Y. Acad. Sci., 111:857 (1964); H. Gottlieb et al., Physiol. and Behavior, 12:61 (1974); T. J. Rosman, J. Pharm. Sci.: 61:46 (1972); D. R. Mishell and M. E. Lumkin, Fert. and Steril., 21:99 (1970). It is possible to determine the amount of drug present in the drugpolymer matrix. F. Theeuwes et al., J. Pharm. Sci., 63:427 (1974). The method of Theeuwes et al. uses differential scanning calorimetry and provides information about drug-polymer interactions in addition to assaying the amount of drug present.